用于病理语音识别的自监督预训练框架研究
Investigating Self-supervised Pretraining Frameworks for Pathological Speech Recognition
摘要：
我们研究了用于自动语音识别（ASR）的病理语音数据集上的自监督预训练框架的性能。现代端到端模型需要数千小时的数据才能很好地进行训练，但只有少量的病理语音数据集是公开的。解决这个问题的一个行之有效的方法是首先在大量健康语音数据集上对模型进行预训练，然后在病理语音数据集上对其进行微调。一种新的预训练框架称为自监督学习（SSL），它只使用语音数据训练网络，在训练数据要求方面提供了更大的灵活性，并允许在预训练中使用更多的语音数据。我们研究SSL框架，如wav2vec 2.0和WavLM模型使用不同的设置，并比较它们的性能与不同的监督预训练设置，使用两种类型的病理语音，即日语electrolaryngeal和英语dysarthric。我们的研究结果表明，虽然SSL已显示出成功的最低限度的资源健康的语音，我们没有发现这是病态的语音的情况。最好的监督设置优于最好的SSL设置的13.9%的字符出错率在喉电语音和16.8%的字出错率在构音障碍的语音。
英文摘要：
We investigate the performance of self-supervised pretraining frameworks on pathological speech datasets used for automatic speech recognition (ASR). Modern end-to-end models require thousands of hours of data to train well, but only a small number of pathological speech datasets are publicly available. A proven solution to this problem is by first pretraining the model on a huge number of healthy speech datasets and then fine-tuning it on the pathological speech datasets. One new pretraining framework called self-supervised learning (SSL) trains a network using only speech data, providing more flexibility in training data requirements and allowing more speech data to be used in pretraining. We investigate SSL frameworks such as the wav2vec 2.0 and WavLM models using different setups and compare their performance with different supervised pretraining setups, using two types of pathological speech, namely, Japanese electrolaryngeal and English dysarthric. Our results show that although SSL has shown success with minimally resourced healthy speech, we do not find this to be the case with pathological speech. The best supervised setup outperforms the best SSL setup by 13.9% character error rate in electrolaryngeal speech and 16.8% word error rate in dysarthric speech.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
